Our long-term goal is to find out how axons in the vertebrate brain determine when and where to form synapses. We are focusing on laminar specificity, a fundamental determinant of connectivity throughout the brain, whereby axons confine their terminal arbors and synapses to specific laminae within a target area. Our experimental object of study is the retinal ganglion cell (RGC), because it is relatively accessible and displays exquisite laminar specificity: axons of distinct RGC subsets synapse in specific sublaminae of the optic tectum's retinorecipient lamina (RRL), and dendrites of distinct RGC subsets arborize in specific sublaminae of the inner plexiform layer (IPL), where they receive inputs from lamina-specified subsets of amacrine and bipolar cells. In work supported by this grant, we have identified several molecules likely to regulate laminar selectivity in this system, and now propose to elucidate their roles: (i) We showed previously that glycoconjugates recognized by a lectin (VVA-B4) are concentrated in the tectal RRL and essential for arborization of retinal axons in that layer. We have now found that versican, a chondroitin sulfate proteoglycan, is the major VVA-binding moiety in this layer. We will ask how it affects retinal axons in vivo and in vitro. (ii) We showed previously that N-cadherin is expressed by all RGCs and promotes RGC arborization in the RRL. We have now found that two Type II cadherins (7 and 11) are expressed by lamina specified RGC subsets and by neuronal subsets in the RRL. We will test the idea that cadherins play multiple roles, with N stabilizing all RGC axons in the RRL and Type II cadherins targeting subsets to RRL sublaminae. (iii) A family of approximately 60 protocadherins, members of which are expressed by RGC subsets, has been proposed to mediate synaptic specificity. We demonstrated synaptic defects in mice lacking all 22 genes of the protocadherin-gamma cluster, but analysis was complicated by widespread neuronal apoptosis in the mutants. We have now been able to avert apoptosis while retaining synaptic defects, so we can analyze synaptic roles directly. (iv) We recently discovered two novel immunoglobulin superfamily adhesion molecules, Sidekicks-1 and -2, and showed that they are critical determinants of sublaminar specificity in the IPL of chicks. We will now use loss- and gain-of-function methods to test their roles in mammals. (v) Finally, we will complete ongoing screens aimed at identifying new markers of RGC subsets in mice, which could be used to follow the development of lamina-specific projections in vivo. Using these strategies, we hope eventually to elucidate mechanisms that promote lamina-specific synapse formation and, by extension, synaptic specificity generally